Operations and Supply Chain Management PDF

OPERATIONS AND SUPPLY CHAIN MANAGEMENT 1. Strategic operations decisions a. INTRODUCTORY CONCEPTS GOODS-SERVICE CONTINUUM Goods are tangible, while services are not, or at least, not so much. E.g. Steel making creates goods. While Teaching provides a service. WHAT IS OPERATIONS MANAGEMENT...

OPERATIONS AND SUPPLY CHAIN MANAGEMENT 1. Strategic operations decisions a. INTRODUCTORY CONCEPTS GOODS-SERVICE CONTINUUM Goods are tangible, while services are not, or at least, not so much. E.g. Steel making creates goods. While Teaching provides a service. WHAT IS OPERATIONS MANAGEMENT? Operations Management is an area of business concerned with the production of goods and services and involves the responsibility of ensuring that business operations are efficient in terms of using as little resource as needed, and effective in terms of meeting customers’ requirements. It’s an area that makes sure supply matches demand. It’s in charge of the management of systems or processes that create goods and/or provide services: SUPPLY CHAIN MANAGEMENT The Supply Chain is the sequence of activities and organizations involves in producing and delivering a good or service. MEASURES TO TAKE DECISIONS - Absolute measures: Revenues, Costs, Operating Income (- Operating Expenses), Net Income (- Interest & Taxes).COSTS: Total costs = FC + VC → TC = FC + Q*VC Fixed costs → do not change with quantity Variable costs → change with quantity Average costs = TC / Q Costs can be seen as a competitive measure when comparing them to several aspects such as: Cost vs. quality The need to produce more with a new technology, a more qualified one. Production must be over Q* because it justifies when the new technology becomes cheaper. Cost vs. variety Variety exists because of: - Market segmentation: more potential customers = more sales. - Competition: allows to compete with others and get more market share and sales - Technology. - International Differences: different markets, different consumers, different preferences. - Government requirements: Different countries, different laws to comply with. For determining whether variety is good or bad we should analyse whether it compensates the Manufacturing Costs vs Customers Choices because: HIGH VARIETY → Higher Complexity Cost (more parts, changeovers…) → Happier Customers (more choices) There will be times in which trade-offs must be made in order to achieve a more optimal production, offer and also to increase sales. This trade-offs imply shifting from one situation to another more optimal.  Cost vs. time Time vs. quality - Relative measures: ROI: Return on Investment OPERATING ROE: Return on Equity ASSET TURNOVER MARGIN ROA: Return on Assets* 𝑂𝑝𝑒𝑟𝑎𝑡𝑖𝑛𝑔 𝑃𝑟𝑜𝑓𝑖𝑡 𝑆𝑎𝑙𝑒𝑠 − 𝐶𝑜𝑠𝑡𝑠 𝑆𝑎𝑙𝑒𝑠 − 𝐶𝑜𝑠𝑡𝑠 𝑆𝑎𝑙𝑒𝑠 𝑆𝑎𝑙𝑒𝑠 𝑆𝑎𝑙𝑒𝑠 − 𝐶𝑜𝑠𝑡𝑠 *𝑅𝑂𝐴 = 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 = 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 = 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠. 𝑆𝑎𝑙𝑒𝑠 = 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠. 𝑆𝑎𝑙𝑒𝑠 This ratios are not enough for understanding the exact impact of operations on assets and sales. Because: - Aggregate measure: it is hard to separate the impact - No details - Direct connection is not transparent - There are many dependencies - External factors. - Survival measures: Cash Flow: difference between a company’s receipts and payments in a given period. b. PRODUCTION STRATEGIES Make To Order (MTO): Consists of beginning production only after receiving an order. Lower Response Time. More Raw Materials than finished goods. Higher Labour Skill level. More general purpose Type of Equipment. Assemble To Order (ATO): Consists of building modules prior to receiving an order so that when an order is received the product is assembled. Higher Response Time. Inventory between Raw materials and finished goods. Mid Labour Skill level. Variety of Equipment. Make To Stock (MTS): Consists of manufacturing the entire product prior to receiving an order and keeping it until an order is placed and the product sent. Higher Response Time. Inventory of finished goods. Low Labour Skill level. Dedicated Type of Equipment. c. TYPES OF PROCESS Work Center / Job Shop: Consists of a collection of general purposes processing resources, together with a highly structured information processing system which allows those resources to be continuously recombined in the production of different outputs. Product is one of a kind and Volume is Low. E.g. Luxury car Manufacturing Cell / Batch Flow: Consists of somewhat between standardized work center – something intermediate between a work center (job shop) and an assembly line. Similar equipment or functions are grouped together. Work frequently is processed in batches. Multiple products but Volume is Low. E.g. Special edition of a product Assembly Line (Line Flow): Consists of a process with Fixed resources, typically specialized or narrow in function, through which work flows in serial fashion. The system is balanced for a particular overall flow rate and a particular product mix, and only small variations around those nominal values are tolerable. Few major products and Volume is relatively High. E.g. Not luxurious cars Continuous Flow: Consists of producing an even more standardized product (vis a vis assembly line) where the product is not a number of discrete parts but rather a continuous flow of undifferentiated product. High standardization of products and quantities and Volume is High. E.g. A bolt factory 2. Project Management PROJECT CHARACTERISTICS (like building construction or a research project) - Job is unique or somewhat familiar - Many related activities - Temporary but critical to the organization - Cuts across organizational lines - The job requires high labour skills MANAGEMENT OF PROJECTS 1. Planning: goal setting, defining the project, team organization. Also includes: o Time/cost estimates o Budgets o Engineering diagrams o Cash flow charts o Material availability details 2. Scheduling: relate people, money, and supplies to specific activities and activities to each other. Also includes: o CPM/PERT o Gantt charts o Milestone charts o Cash flow schedules 3. Controlling: monitor resources, costs, quality, and budgets; revise plans and shift resources to meet time and cost demands. Also includes: o Budgets o Delayed activities report o Slack activities report THE ROLE OF THE PROJECT MANAGER Responsible for making sure that: - All necessary activities are finished in order and on time - The project comes in within budget - The project meets quality goals - The people assigned to the project receive motivation, direction, and information Plus must be: Good coaches, good communicators, and able to organize activities from a variety of disciplines. ETHICAL ISSUES PROJECT MANAGERS MAY FACE Project Management Institute has established an ethical code to deal with problems such as: - Offers of gifts from contractors - Pressure to alter status reports to mask delays - False reports for charges of time and expenses - Pressure to compromise quality to meet schedules WORK BREAKDOWN STRUCTURE PROJECT  MAJOR SUBTASKS IN PROJECT SUBTASKS IN MAJOR TASKS ACTIVITIES / “WORK PACKAGES” TO BE COMPLETED PROJECT SCHEDULING TECHNIQUES: Gantt chart, CPM, PERT 1. GANTT CHART: They can be made with several project management software like: Oracle Primavera, MindView, HP Project, Fast Track, Microsoft Project. CPM AND PERT SHARE 6 CRUCIAL STEPS. They are: a. Define project and prepare the work breakdown structure (mentioned above). b. Develop relationships among the activities (decide which must precede and which follow others). c. Draw the network connecting all of the activities. d. Assign time and/or cost estimates to each activity. e. Compute the longest path through the network (critical path). f. Use the network to help, plan, schedule, monitor, and control the project. 2. CRITICAL PATH METHOD (CPM): It is a network technique created by DuPont in 1957 for chemical plants. CPM assumes we know a fixed time estimate for each activity and there is no variability in activity times 3. PROGRAM EVALUATION AND REVIEW TECHNIQUES (PERT): It is a network technique created by Booz, Allen & Hamilton in 1958 with the US Navy for Polaris missile. PERT uses a probability distribution model for activity times to allow for variability. COMPARISON OF “AON” AND “AOA” NETWORK CONVENTIONS DETERMINING THE “PROJECT SCHEDULE” Performing a Critical Path Analysis (CPM) - Critical Path is the longest path through the network. - Critical Path is the shortest time in which the project can be completed. - Any delay in Critical Path Activities delays the project. - Critical Path Activities have no slack time. Example of network for getting critical path: A (2 weeks), B(3), C(2), D(4), E(4), F(3), G(5), H(2) → CP = ACEGH = 15 weeks Possible Paths: ACFH → 9 weeks ACEGH → 15 weeks → CP ADGH → 13 weeks BDGH → 14 weeks Once we’ve figured out the CP we can then start getting more information such as: Earliest Start (ES) → earliest start at which an activity can start, assuming all predecessors have been completed. ES = Max {EF of all immediate predecessors} Earliest Finish (EF) → earliest time at which an activity can be finished. EF = ES + Activity time Latest start (LS) → latest time at which an activity can start so as not to delay the completion time of the entire project. LS = LF – Activity time Latest Finish (LF) → latest time by which an activity has to be finished so as to not delay the completion time of the entire project. LF = Min {LS of all immediate following activities}  Slack time (ST) → It is the free time for each activity: The length of time an activity can be delayed without delaying the entire project. It can be computed after having got the EF, ES, LF, and LS times for all activities. ST = LS – ES or ST = LF – EF Activities on Critical Path have no slack time: ST = 0 STEPS FOR CPM ANALYSIS: 1º Draw network (either AON or AOA) 2º Determine the possible paths and their times 3º Choose the Critical Path CP 4º Draw table and get ES and EF 5º Compute Slack Time from ST = LF – EF or ST = LS – ES. You must take into account the possible paths for each activity and choose the one of them with the highest time, then: ST = CP length – chosen path. 6º Compute LF and LS from isolating them from ST = LF – EF or ST = LS – ES 7º (optional) Redraw network as: // Update table: Analysis can be a Forward or Backward Pass. Finally, we can plot all these data into Gantt charts: LS – LF Gantt chart ES – EF Gantt chart Performing a Project Evaluation and Review Techniques As previously mentioned, PERT uses a probability distribution model for activity times to allow for variability. Then, we must first take into account that three time estimates are required: - Optimistic time (a): if everything goes as planned. - Pessimistic time (b): very unfavourable conditions - Realistic or most likely time (m) The estimate follows beta distribution thus we must get: a) Expected activity time (t): t = (a + 4m + b) / 6 b) Variance of activity completion times (σ2 or v): v = [(b-a) / 6]2 Beta distribution means: STEPS FOR PERT ANALYSIS: 1º From the given table with optimistic (a), realistic (m) and pessimistic (b) times get the Expected time of each activity using t = (a + 4m + b) / 6 Also Variance of each activity using v = [(b – a)]2 2º Draw the network and for times of each activity use the ones you got from step 1º. You now can also compute the Project Variance (σ2p) which is the sum of the variances of activities on the critical path. And the Project Standard Deviation (σp): Square Root of the Project Variance 3º Now you can complete a full table with (a), (m), (b), (t), (v), and now get ES, EF, LS, LF and ST taking into account (t) 4º Since PERT follows a Normal Distribution (0, 1) (µ, σ) (mean, std) for getting the Probability of Project Completion we need to STANDARDIZE the variables so looking for a probability in the Normal Table is possible. Done as follows: 𝑋−µ 𝐷𝑢𝑒 𝑑𝑎𝑡𝑒 – 𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑑𝑎𝑡𝑒 𝑜𝑓 𝑐𝑜𝑚𝑝𝑒𝑡𝑖𝑜𝑛 (𝑪𝑷 𝑻𝑰𝑴𝑬) 𝑃(𝑍 < 𝑥) → 𝑃 ( < ) σ σ Z is the number of STDs the due date or target date lies from the mean or expected mean. 5º Look in Normal Distribution table for the value we just got in 4º and x100 for % 6º We now know expected completion time, % chance of getting project done before X, activities on CP, activities not on CP. Now we are also able to get info into Gantt charts. Variability of times for activities on noncritical paths must be considered when finding the We are not probability of finishing in a specified time. including this Variation in noncritical activity may cause change in critical path consideration in our probability calculations, but it is something to keep in mind, especially for those noncritical activities with large variances.  BUDGET CONTROL Very important aspect of project management. Costs can exceed budget due to overly optimistic time estimates or unforeseen events. Unless corrective action is taken, serious costs overruns can occur. TIME-COST TRADE-OFFS: CRASHING Activity time estimates are made for some given level of resources. But it may be possible to reduce the duration of a project by injecting additional resources. What for? → - Avoiding penalties - Monetary incentives or interest - Free resources for use on other projects - Project behind schedule - Completion time has been moved forward Crashing means shortening activity durations, and hence, project’s duration. Normally by adding more funds for more equipment or personnel. By directly increasing expenses but reducing times, savings in indirect costs are made. Factors to consider when crashing a project: - The amount by which an activity is crashed is, in fact, permissible. - Taken together, the shortened activity durations will enable us to finish the project by the due date. - The total cost of crashing is as small as possible. STEPS IN PROJECT CRASHING: Data regarding costs and time reduction of each crash should be given. 1º Compute the crash cost per time period. If crash costs are linear over time: (𝐶𝑟𝑎𝑠ℎ 𝑐𝑜𝑠𝑡 − 𝑁𝑜𝑟𝑚𝑎𝑙 𝑐𝑜𝑠𝑡) CHECK PHYSICAL // HAND MADE 𝐶𝑟𝑎𝑠ℎ 𝑐𝑜𝑠𝑡 𝑝𝑒𝑟 𝑝𝑒𝑟𝑖𝑜𝑑 = (𝑁𝑜𝑟𝑚𝑎𝑙 𝑡𝑖𝑚𝑒 − 𝐶𝑟𝑎𝑠ℎ 𝑡𝑖𝑚𝑒) EXAMPLE OF “MILWAUKEE” 2º Using current activity times, find the critical path(s) and identify the critical activities. 3º If there is only one critical path, then select the activity on this critical path that (a) can still be crashed, and (b) has the smallest crash cost per period. If there is more than one critical path, then select one activity from each critical path such that (a) each selected activity can still be crashed, and (b) the total crash cost of all selected activities is the smallest. Crash each selected activity by one period. 4º Update all activity times. If the desired due date has been reached, stop. If not, return to Step 2. ADVANTAGES OF PERT AND CPM: - Especially useful when scheduling and controlling large projects - Straightforward concept and not mathematically complex - Graphical networks help highlight relationships among project activities - Critical Path and Slack Time analyses help pinpoint activities that need to be closely watched - Project documentation and graphics point out who is responsible for various activities - Applicable to a wide variety of projects - Useful in monitoring not only schedules but costs as well LIMITATIONS / DISADVANTAGES OF PERT AND CPM: - Project activities have to be clearly defined, independent, and stable in their relationships - Precedence relationships must be specified and networked together - Time estimates tend to be subjective and are subject to fudging by managers - There is an inherent danger of too much emphasis being placed on the longest, or critical, path 3. Forecasting CHARACTERISTICS OF FORECASTING - Forecasts are always inaccurate - Long-term forecasts are usually less accurate than short-term forecasts - Aggregate forecasts are usually more accurate than disaggregate forecasts - In general, the farther up the supply chain a company is, the greater is the distortion of information it receives COMPONENTS AND METHODS Companies must identify the different factors that can affect them (and that may not be common with other competitors) and the influence of each of them to ascertain the relationship between them and future demand. Some of them are: Past demand, Planned price discounts, State of economy, competitors’ actions, marketing efforts… 1. Time series: Only use historical demand. They are better with stable demands. With time series methods we must know the COMPONENTS OF AN OBSERVATION: Observed demand (O) = systematic component (S)* + random component (R)** - * Systematic component (S): expected value of demand: - Level: current deseasonalized demand - Trend: growth or decline in demand - Seasonality: predictable seasonal fluctuation - ** Random component (R): part of the forecast that deviates from systematic component. - Forecast error: Difference between Forecast and Actual Demand There can be different BEHAVIOURS: - Level (1) → No Pattern - Trend (2) → Linear (or default) or Nonlinear - Seasonality (3) → Repetition at Fixed Intervals (1) (2) (3) COMPONENTS OF DEMAND There are several Time series models, different forecasting methods with different applicability: - Moving average: applicable when there is NO Trend or Seasonality - Exponential smoothing: applicable when there is NO Trend or Seasonality - Holt’s model: applicable when THERE IS Trend but NO Seasonality - Winter’s model: applicable when THERE IS Trend AND Seasonality STEPS IN FORECASTING: 1º Initialize: Compute initial estimates of level (L0), Trend (T0), and Seasonal Factors (S1,…,Sp) 2º Forecast: Forecast demand for period t + 1 (Ft + 1) 3º Revise estimates (adaptive): Modify the estimates of Level (Lt + 1) and Trend (Tt + 1) and Seasonal Factors (St + p + 1) 4º Estimate error: Compute error Et + 1 = Ft + 1 – Dt + 1 MOVING AVERAGE (MA) (Time series models method) Remind: NO Trend, NO Seasonality. Thus, systematic component of demand (S) = Level 1º Initialize: Level in period t is the average demand over the last N periods: (𝐷𝑡 + 𝐷𝑡−1 + … + 𝐷𝑡−𝑁+1 ) 𝐿𝑡 = 𝑁 2º Forecast: Forecast in period t + 1 is: 𝐹𝑡+1 = 𝐿𝑡 3º Revise estimates: After observing demand for period t + 1, revise the estimates: (𝐷𝑡+1 + 𝐷𝑡 + … + 𝐷𝑡−𝑁+2 ) 𝐿𝑡+1 = 𝑁 𝐹𝑡+2 = 𝐿𝑡+1 4º Estimate error: Et + 1 = Ft + 1 – Dt + 1 + *See MEASURES OF ERROR BELOW* With MA data uses N periods, they all have the same weight. The larger the N, the more stable the model is, but less responsive to most recently observed demand. EXPONENTIAL SMOOTHING (Time series model method) Remind: NO Trend, NO Seasonality. Thus systematic component (S) = Level 1º Initialize: Level in period t is: 𝐿𝑡 = 𝛼. 𝐷𝑡 + (1 − 𝛼). 𝐹𝑡 where L0 = D0 2º Forecast: Forecast for period t + 1 is: Ft + 1 = Lt 3º Revise estimates: After observing demand for period t + 1, revise the estimates: 𝐹𝑡+2 = 𝐿𝑡+1 = 𝛼. 𝐷𝑡+1 + (1 − 𝛼). 𝐹𝑡+1 4º Estimate error: Et + 1 = Ft + 1 – Dt + 1 + *See MEASURES OF ERROR BELOW* This model is linked to MA as for getting L0 we must compute the average demand over the last N periods. We use data of all periods we have. Most recent data has more weight. As α increases, the process is more responsive to changes in recent demand.  TREND – CORRECTED EXPONENTIAL SMOOTHING (HOLT’S MODEL) (Time series model method) Remind: NO Seasonality, but YES Trend. Thus systematic component (S) = Level + Trend 1º Initialize: Obtain initial estimate of Level and Trend by running a linear regression 𝐷𝑡 = 𝑎𝑡 + 𝑏 𝑇0 = 𝑎 𝐿0 = 𝑏 2º Forecast: In period t the forecast for future periods is: 𝐹𝑡+1 = 𝐿𝑡 + 𝑇𝑡 and 𝐹𝑡+𝑛 = 𝐿𝑡 + 𝑛. 𝑇𝑡 3º Revise estimates: After observing demand of time t + 1, revise: 𝐿𝑡+1 = 𝛼. 𝐷𝑡+1 + (1 − 𝛼). (𝐿𝑡 + 𝑇𝑡 ) and 𝑇𝑡+1 = 𝛽. (𝐿𝑡+1 − 𝐿𝑡 ) + (1 − 𝛽). 𝑇𝑡 4º Estimate error: Et + 1 = Ft + 1 – Dt + 1 + *See MEASURES OF ERROR BELOW* TREND AND SEASONALITY – CORRECTED EXPONENTIAL SMOOTHING (WINTER’S MODEL) (Time series model) Remind: YES Seasonality, YES Trend. Thus systematic component (S) = (Level +Trend). seasonal factor 1º/2º Initialize / Forecast: F1 = (L0 + T0). S1 Ft + i = (Lt + i. Tt). St + i i=1,…,p 3º Revise estimates: After observing demand for period t + 1, revise estimates for Level, Trend and Seasonal factors: 𝐷 𝐿𝑡+1 = 𝛼. ( 𝑆𝑡+1) + (1 − 𝛼). (𝐿𝑡 + 𝑇𝑡 ) 𝑡+1 𝑇𝑡+1 = 𝛽. (𝐿𝑡+1 − 𝐿𝑡 ) + (1 − 𝛽). 𝑇𝑡 𝐷𝑡+1 𝑆𝑡+𝑝+1 = 𝛾. ( ) + (1 − 𝛾). 𝑆𝑡+1 𝐿𝑡+1 Where α = constant for Level, β = constant for Trend, γ = constant for seasonal factor. 4º Error: Et + 1 = Ft + 1 – Dt + 1 + *See MEASURES OF ERROR BELOW* MEASURES OF FORECAST ERROR - Error: Et = Ft – D t - Bias: Not absolute 𝑏𝑖𝑎𝑠𝑛 = ∑𝑛𝑡=1 𝐸𝑡 - Mean absolute deviation (MAD): Sum of the errors in absolute value divided by N 1 𝑀𝐴𝐷𝑛 =. ∑𝑛𝑡=1 𝐴𝑡 , At = |Et| 𝑛 - Tracking Signal (TS): TS = Bias / MAD Capacity:  o Curtail demand by raising prices, scheduling longer lead times, discouraging marginally profitable business. o Long – term solution is to increase capacity IF Capacity > Demand: o Stimulate demand through price reductions or aggressive marketing o Product changes Adjusting to seasonal demands: o Produce products with complementary demand patterns 2. Manage supply using capacity, inventory, subcontracting, and backlogs. Managing Capacity: o Time flexibility from workforce o Use of seasonal workforce o Use of subcontracting o Use of dual facilities – specialized and flexible o Designing product flexibility into production processes Managing Inventory: o Build inventory of high demand or predictable demand products o Using common components across multiple products o Exploiting substitution o Postponing product differentiation (delayed differentiation) INVENTORY / CAPACITY TRADEOFF Levelling capacity forces inventory to build up in anticipation of seasonal variation in demand. While carrying low levels of inventory requires capacity to vary with seasonal variation in demand or enough capacity to cover peak demand during season. POSTPONEMENT Delay product differentiation or customization until closer to the time the product is sold, it helps responding to predictable variability and be better prepared in case of seasonal changes and more flexibility. How? → - Have common components in the supply chain for most of the push phase* - Move differentiation as close to the pull phase** of the supply chain as possible - Inventories in the supply chain are mostly aggregate * Push face of supply chain: earlier stage in the supply chain where product is still almost not done, and it is long-term driven. ** Pull phase of supply chain: latter phase close to sale of the product where some modifications can still be made for the client. RISK POOLING STRATEGIES Idea: Pooling separate sources of demand reduces uncertainty Four versions of risk pooling: Location pooling: puesta en común de la ubicación A single location stores inventory used by several sales reps. Sales reps no longer hold their own inventory, they must pull inventory from the pooled location. Inventory is automatically replenished at the pooled location as depleted by demand. Pros: Reduces demand uncertainty, which allows a firm to reduce inventory, increase service, or a combination of both. Cons: Location pooling moves inventory away from customers Allocation issues Alternative: Virtual pooling: each sales reps. Keep own inventory but shares inventory with nearby reps only if needed. Product pooling (or universal design): mancomunación de productos (diseño universal) Lead time pooling (or delayed differentiation): puesta en común del tiempo de entrega Capacity pooling: puesta en común de la capacidad Four possible capacity configurations: no flexibility to total flexibility Flexibility allows production shifts to high selling products to avoid lost sales. Adding flexibility increases capacity utilization and expected sales. 20 links can provide nearly the same performance as total flexibility Example: 2 plant, 2 product example and 2 configurations: no flexibility and total flexibility →  Demand 75 for P.A and 115 for P.B then: SUMMARY OF RISK POOLING The main idea is to reduce demand uncertainty by aggregating demands (of different location, products). The relationship between demands: - Pooling is useful for negatively correlated or independent demands - Pooling does not yield any benefit for positively correlated demands 6. Statistical Process Control a. STATISTICAL PROCESS CONTROL (SPC) Statistical Process Control is the application of the statistical techniques to ensure that processes meet standards. Variability: variation is inherent in any process: - Random variation (common causes): natural variation in output of a process, created by countless minor facts. - Assignable variation (special causes): a variation whose source can be identified. The objective of a process control system is to provide a statistical signal when assignable causes are present. Therefore, it monitors quality of manufacturing and service processes to identify and eliminate assignable variation. TYPES OF VARIATIONS SCENARIOS - In Control: o No special causes (no assignable variation) o Easy to make predictions o Our output specifications are stable - Out of Control: o Special causes are present o It is not possible to make predictions o Unstable output To determine whether the process is in control: CONTROL CHARTS b. CONTROL CHARTS The purpose of control charts is to help distinguish between natural variations and variations due to assignable causes. TYPES OF CHARTS - Variable Control Charts (Continuous metric) o X-bar chart → shows changes in relation to the mean o R chart → shows gain or loss in terms of dispersion - Attribute Control Charts (Discrete metric) o P-chart → those that measure the percent defective in a sample o C-chart → those that count the number of defects They all follow the same principle (Center Line (CL), Upper Control Limit (UCL) and Lower Control Line (LCL)) X-BAR CHART – CENTRAL LIMIT THEOREM AND R CHART Central Limit Theorem states that regardless of the distribution of the population, the distribution of means (each of which is a mean of a sample drawn from the population) will tend to follow a normal distribution. The mean of the averages will equal the mean of the overall population: ̅= 𝑥 µ CENTER LINE (CL) = ̅ 𝑥 FOR X-bar Chart  UPPER CONTROL LIMIT (UCL) = ̅ 𝑥 + A2 *R ̅ – A2 *R LOWER CONTROL LIMIT (LCL) = 𝑥 CENTER LINE (CL) = R UPPER CONTROL LIMIT (UCL) = D4 *R FOR R Chart LOWER CONTROL LIMIT (LCL) = D3 *R Both charts are needed because: The normal distribution is defined by two parameters, the mean and the standard deviation. The mean (mean chart) and the R-Chart mimic these two parameters. The X-chart is sensitive to shifts in the process mean, whereas the R-chart is sensitive to shifts in the process standard deviation. How to obtain stable control limits? 1. Collect m samples of size n. 2. Calculate ̅ 𝑥 ,R, and then: CL, UCL, and LCL from the sample 3. Plot points: If all the points are in the control range, the limits will be stable control limits and the job is done. If not, the limits become trial control limits. Go to step 4. 4. Find out the causes and fix if they are manageable. 5. Exclude the outlier sample and go back to step 2. MIRAR EJEMPLO DE LAS SLIDES, 19-29 HOW TO TELL IF THE PROCESS IS NOT IN CONTROL? - Non-random patterns: Trend, Cycles, Bias, Mean shift, Too much dispersion - Rules: o Point beyond control limits (special cause) o Eight or more consecutive points on one side of the line o Six or more consecutive points increasing or decreasing o Fourteen or more consecutive points oscillating CHARTS FOR ATTRIBUTES: p-CHART and c-CHART - P-Chart: Control chart used to monitor the proportion of non-conforming items in a process (Use when the data can be converted into proportions). Ex: proportion of patients admitted in less than 4 hours - C-Chart: Control chart used to monitor the number of non-conformances (defects) per unit (Use when the number of occurrences per unit of measure can be counted; non-occurrences cannot be counted). Ex: number of service tickets closed in a day. c. PROCESS CAPABILITY Control charts are used because they are proven, effective at defective prevention, and effective at preventing unnecessary correction. PROCESS CAPABILITY vs. IN CONTROL “In statistical control” refers to a process which is consistently producing parts subject to the same random variation (i.e. there is no detectable special cause variation). Conformance to specifications, in contrast, means that we are producing parts which meet a certain standard. Suppose we are making a component using a certain process. Our internal standard (or the customer’s) requires that we produce at least 99.5% good components. No matter what we do, we find that we can produce only 98% good components. Then we say that our process is not capable. Suppose we find that we can produce 99.5% (or more) acceptable products i.e. the process is capable. Due to many reasons (such as worn out tool), the process may not perform up to expectations at a particular time. Here are two questions we want to answer. 1. Is the process capable? To answer this, we express it in a measure called “Process Capability Ratio, (Cp)” 2. Even if the process is capable, what is the current status? To answer this, we use the “Process Capability Index, (Cpk)” PROCESS CAPABILITY RATIO 𝑈𝑆𝐿 − 𝐿𝑆𝐿 Process Capability Ratio (Cp) = , 6𝜎 where USL = Upper Specification Limit. And LSL = Lower Specification Limit Ex: Stable process, normally distributed. Critical dimension specifications 2.500 ± 3”. Estimated standard deviation = 1”. Calculate the value of the process capability ratio. Cp = (2.503 – 2.497) / (6*1) = 1  INTERPRETATION OF THE RATIO Process Capability Ratio gives us the probability of acceptance and rejection as shown (rejection number is also calculated as parts per million (ppm). As a “rule of thumb” companies strive for Cp = 1,66 IN EXCEL: - Prob. of rejection = 2 * (1 – NORMDIST(3 * Cp)) - Prob. of acceptance = 1 – Prob. of rejection Example: Stable process, normally distributed. Tolerance limits for critical dimension: 2.500 ± 3,5. Estimated std dev = 1 1. Process Capability Ratio or Cp: USL = 2.500 + 3,5 = 2.503,5 LSL = 2.500 – 3,5 = 2.496,5 Cp = (2.503,5 – 2.496,5) / (6 * 1) = 1,167 NOW: Suppose process average shifts to 2.501 as tool wears out. Then we cannot use Process Capability Ratio (Cp), but need to calculate what the process is doing at the moment. That is achieved by getting the Process Capability Index (Cpk). PROCESS CAPABILITY INDEX Process Capability Index (Cpk) = min (Cpu, Cpl) 𝑈𝑆𝐿 − µ µ − 𝐿𝑆𝐿 Where: Cpu = , Cpl = , and µ = actual process mean 3𝜎 3µ Example 2 (cont.): Stable process, normally distributed. Critical dimension - specifications 2.500 ± 3. Measured standard deviation = 1. If average = 2.501 Calculate process capability index. 2. Process Capability Index or Cpk: Cpu = (2.503 – 2.501) / (3 * 1) = 0,667 Cpl = (2.501 – 2.497) / (3 * 1) = 1,333 Then: Cpk = min (Cpu, Cpl ) → Cpk = 0,667 HOW IS THE PROCESS CAPABILITY USED 1. Select target process capability (customer requirement/own standard/industry standard) 2. Calculate Process Capability Ratio (Cp) a. Cp < target value: Cannot use process, need long term solution: better process, train operators, better quality raw material… Need to reduce variability (std. deviation) b. Cp > target value: Use the process and periodically calculate Cpk If Cpk < target value: need simple adjustments (sharpen tool, adjust setting…) 7. Total Quality Management Total Quality Management refers to a style/methodology of management that encompasses the entire organization, from supplier to customer. The aim is to drive the company towards excellence in all aspects of products and services that are important to the customer. How can a company implement Total Quality Management? - Continuous Improvement - Just-in-time (JIT) - Six Sigma - Taguchi concepts - Employee Empowerment - TQM Tools - Benchmarking 1. CONTINUOUS IMPROVEMENT Continuous improvement is a way of working based on a never ending process of seeking perfection that covers people, equipment, suppliers, materials and procedures. That perfection is never achieved, but always sought. How to implement a continuous improvement culture? PDCA: Plan – Do – Check – Act Kaizen concept = “improvement” in Japanese Look for “Zero defects” 2. SIX SIGMA Six Sigma is a program designed to reduce defects to save time, improve quality, lower costs and achieve total customer satisfaction. Statistical definition of a process that is 99,9997% capable, 3.4 defects per million opportunities (DPMO) How can an Operations Manager implement a Six Sigma methodology? 1) Defines the project’s purpose, scope and outputs 2) Measures the process and collects data 3) Analyses the data, ensuring repeatability 4) Improves, by modifying or redesigning, existing process and procedures 5) Controls 3. EMPLOYEE EMPOWERMENT  Employee Empowerment means involving employees in every step of the production process. In other words, it consist of moving part of the responsibility (decisions taken) from the higher and medium layers to the lower levels. It motivates workers. It is also known as Horizontalization. 4. BENCHMARKING Benchmarking involves selecting a demonstrated standard of products, services, costs, or practices that represent a very best performance for processes or activities very similar to your own. Benchmarking can be done internally and externally. How can a Company develop a benchmarking methodology? 1) Determine what to benchmark 2) Form a benchmark team 3) Identify benchmarking partners 4) Collect and analyse benchmarking information 5) Take action to match or exceed the benchmark 5. JUST IN TIME (JIT) Just in time (JIT) is a philosophy based on continuing improvement and enforced problem solving. Therefore, JIT systems are designed to produce or deliver goods just as they are needed. How is JIT related to Quality? - Cutting the cost of quality: This occurs because rework, scrap, inventory investment and damage cost are directly related to inventory on hand. Because there is less inventory on hand with JIT, costs are lower. Inventory also hides bad quality; therefore JIT exposes bad quality. - Improving quality: As JIT shrinks flowtime, it keeps evidence of errors fresh and limits the number of potential sources of error. JIT creates, in effect, an early warning system for quality problems, both within the firm and with vendors. - Reducing inventory cost: As far as JIT processes produce and deliver goods just as they are needed, the inventory cost decreases drastically. 6. TAGUCHI CONCEPTS Taguchi concepts are Engineering and experimental design methods to improve product and process design. It is based on identifying key components and process variables affecting product variation. They are: - Quality robust: Total focus on the robustness of our product in order to meet customer needs. - Target oriented quality: Continuous improvement to get our product to our target quality. - Quality loss function: Mathematical relation between cost and distance to target quality. TOTAL QUALITY MANAGEMENT (TQM) – TOOLS For generating ideas: - Check Sheet: designed for recording data. In many cases, the recording is done so the patterns are easily seen while the data are being taken. It is helpful for identifying patterns that may require further analysis. - Scatter Diagram: they show the relationship between two measurements. Useful for understanding the relation (+ / - ) between the variables and its dispersion. - Cause and effect Diagram: it is a schematic technique used to discover possible locations for quality problems. It could be completed using other techniques, such as brainstorming process. For organising data:  - Pareto Chart: it is a methodology for organising errors, problems or defect to help focus on problem solving efforts. (80% - 20% Rule). - Flowchart: it graphically presents a process or system using annotated boxes and interconnected lines. They are simple but great tools for trying to make sense of a process or explain a process. For identifying problems: - Histograms: they graphically show the range of values of a measurement and the frequency with which each value occurs. - Statistical Process Control (SPC): they monitor standards, make measurements and take corrective actions as products or services are being produced. 8. Supply Chain Management A Supply chain comprises all the parties involved in providing a product or service to a customer, either directly or indirectly. Supply Chain Management is the set of approaches utilized to efficiently integrate suppliers, manufacturers, warehouses, and stores, so that merchandise is produced and distributed: - at the right quantities - to the right locations - at the right time - in order to minimize system-wide costs while satisfying service level requirements SUPPLY CHAIN FLOWS 3 TYPES: Information (1), Product or Material (2), Finance or Funds (3) FLOWS CAN BE UPSTREAM (normally towards manufacturers or even suppliers) OR DOWNSTREAM (normally towards customers) Returns, repairs, Raw materials, servicing, recycling, intermediate products, disposal finished goods Sales, orders, inventory, Capacity, promotion quality, promotion plans, delivery plans schedules Payments, Credits, consignment, consignments payment terms, invoice SUPPLY CHAIN MANAGEMENT (SCM) DECISIONS  - What suppliers should we use and how many? - How many factories and warehouses should we have? - What products should each factory produce? - In what locations should our firm have warehouses and factories? - How do we set capacity at each location? - Given locations and capacities, what quantities should we produce and store at these locations? - What quantities should move from location to location and at what time? DECISION PHASES OF A SUPPLY CHAIN Supply chain STRATEGY or DESIGN: How to structure the supply chain over the next several years: Decisions about the structure of the supply chain and what processes each stage will perform. Strategic supply chain decisions: - Locations and capacities of facilities - Products to be made or stores at various locations - Information systems type Supply chain design must support strategic objectives. Design decisions are long-term and expensive to reverse – must take into account market uncertainty. Supply chain PLANNING: Decisions over the next quarter or year: Definition of a set of policies that govern short-term operations. Fixed by the supply configuration from previous phase. It starts with a forecast of demand in the coming year. Planning decisions: - Planned build-up of inventories - Subcontracting, backup locations - Inventory policies - Timing and size of market promotions Must consider in planning decisions demand uncertainty, exchange rates, competition over the time horizon. Supply chain OPERATION: Daily or weekly operational decisions: Time horizon is weekly or daily. Decisions regarding individual customer orders Supply chain configuration is fixed, and operating policies are determined Goal is to implement the operating policies as effectively as possible. Operational decisions: - Set order due dates - Generate pick lists at a warehouse - Allocate an order to a particular shipment - Set delivery schedules Much less uncertainty (short time horizon). PUSH / PULL VIEW OF A SUPPLY CHAIN Supply chain processes fall into one of two categories depending on the timing of their execution relative to customer demand: Pull: execution is initiated in response to a customer order (reactive) Push: execution is initiated in anticipation of customer orders (speculative) Push/pull boundary separates push processes from pull processes: DELL AS AN EXAMPLE OF BEST PRACTICE: Direct business model: - No middlemen - No warehouses - Suppliers are close by Build-to-order strategy: - Production doesn’t begin until an order is placed - There is never more than 4 hours inventory on hand - Leads to lower cost They have a global supply chain that gets components to Austin, Texas, where assembly is done. Before 2008: Focus on responsiveness In 2008: DELL enters retail channel → supply chain setup does not support low cost (online customers are less price-sensitive than retail customers). There is a need to re-engineer, and they had to segment supply chain using: - Demand uncertainty - Cost drivers - Relationships with customers - Customer value proposition - Technology clock-speed Customer driven supply chain segmentation: Cost-responsiveness efficient frontier: ZARA AS AN EXAMPLE OF BEST PRACTICE: Success due to:  - Stockout often - Have shops in the most expensive parts of the world - Redesigns apparel all time, which lead to demand uncertainty - Does not advertise - No product innovation - No technology innovation - Not a new niche Innovation through better supply chain management. Transferred DELL’s build-to-order model to the apparel industry. Fashion-to-order: - Close production location, short lead times - Quantity decisions made days before sales - Detailed product differentiation decision is postponed - Few days of inventory - 21-29 day cycle Pays with more expensive capacity but the trade-off is favourable STRATEGIC FIT MATCHING SUPPLY CHAIN WITH PRODUCTS Match Mismatch Mismatch Match Functional → Certain // Innovative → Uncertain FACILITIES INVENTORY TRANSPORTATION / SHIPPING SOURCING PRICING Every-Day-Low-Pricing EFFICIENT / RESPONSIVE SUPPLY CHAINS 8.2 Supply Chain Design HIERARCHY OF NETWORK DESIGN DECISIONS Network Regional Selecting Location Design Facility Desirable Sites choices Strategy Configuration Stage I: Network Design Strategy The goal is to identify the broad structure of the supply chain network based on business strategy, customer segments served, and product characteristics. Decide whether it is done in-house/outsourced, how many layers/tiers?, products, markets… As well as what factors influencing network design decisions: - Strategic (Customer needs): Response time Customer experience Product variety Order visibility Product availability Returnability - Technological - Macroeconomic - Political - Infrastructure - Competitive - Logistics and facility costs What are alternative structures for a distribution network? How do they compare in terms of supply chain costs and capabilities? Which customers and products are they suited for? The relation between Service and Number of facilities is: The more facilities, the higher response time. The relation of the Costs as a Function of Number of facilities is: Stage II: Facility Configuration We must ask ourselves, Which regions should have facilities? And analyse the key factors: - Regional demand configuration - Production cost - Taxes and tariffs - Inventory cost - Fixed facility costs (economies of scale) - Coordination cost  - Transportation cost Demand per region, capacity of each facility…must be figured out. Stage III: Facility Location, Capacity and Demand Allocation Typical problem after a merger is rationalizing supply chain network. Questions such as: - Which plants to keep open? - What capacity to assign to each plant? - Which markets to assign to each plant? Multi-stage location and allocation decisions: - Which plants to establish? - Which plants to distribute to which markets? - How to configure the network? Structure of an optimization problem: - Parameters (know values a priori) - Decision variables (unknown, we want to find this out) 9.1 Circular Economies and Green Supply Chain Reverse logistics process is the one a returned product goes through. From wholesalers or retailers back to the producers. It involves several stages such as: Acquisition of used (returned products) → Logistics process → Disposition and remanufacturing process → Remarketing and reselling Any firm having returned products must find out the causes of the devolutions in order to amend them, as well as learn how to leverage value of unavoidable returns (resell, scrap, recycle…). Returns can come from Consumer and Channel (retailers, wholesalers…). HOW TO REDUCE RETURNS? 1. Reduce Root Cause of returns: carry on initiatives in order to reduce the # of returns 2. Reduce cost per unit shipped due to returns through improved management of the product returns process (reverse logistics) There can be several reasons why a product is returned. Some may be: - Over stocks - Lower prices - Stock adjusting - Defective product - Perception on Customer expectations - Rentals - Unfulfilled expectations RETURNS CAN BE SEEN FROM 2 DIFFERENT PERSPECTIVES: 1. “Necessary Evil” (A bad view, something there is no remedy for), from a Marketing approach. Treats returns as a consequence of sales, and as an opportunity for new sales. 2. Business unit. Treats it as another revenue section of the company that acts as an individual company within the actual company. This approach requires a much more detailed analysis. Some aspect this analysis must take into account are: - Strategic partnerships and alliances (outsource processes possibility) - Processes related to remanufacture - Return on Investment (ROI) - Reverse logistic structure - Cost of capital - Activity costs Decentralized returns supply chain are bound to result chaotic and to increase times which is not ideal. An example of decentralized returns supply chain would be: Whereas a Centralized returns supply chain helps to lower the amount of value that is lost and maximize value recovery. Example: Return process must be easy for those who return products i.e. wholesaler, consumer, retailer. HOW TO LEVERAGE VALUE FROM UNAVOIDABLE RETURNS? Through a Closed-Loop supply chain. This type of supply chain contribute to maximize value recovery, reduce overall loss of value, and leverage value as after an analysis a returned product can have several uses. For instance, to be disassembled and become scrap, to be recycled, to be remanufactured and resold or to reuse components for new products. Having common components across different product lines helps increase even more the value recovery as a part that will be reused can have several ends rather than just one or be thrown away. Closed-Loop Supply Chains attempt to integrate the forward flow and the different return flows into a single service process to the customer. DOES IT MAKE SENSE TO ENGAGE IN REMANUFACTURING ACTIVITIES? To decide if it is worth doing the remanufacturing process in-house rather than outsourcing it, businesses must assess the pros and cons. - Pros of outsourcing: faster (allows to remarket in less time), cheaper - Cons of outsourcing: loss of control, dependency on third-parties, reveals a lo t about how the product is done (therefore, risks of leaks), job may not be done as good as in- house or at least not properly 100%. Furthermore, companies must take into account the previously mentioned ROI and cost of capital, if cost of capital is bigger than earnings of this “business unit”, they should reconsider changing outsource provider or stop in-house remanufacture (depending on which one is being done). Look for the highest ROI possible with decent quality standards. What can be done to maximize value recovery? - Improve velocity of reverse flow by: express shipments, faster processing at Testing & Refurbishing facilities, redesign.. - Outsource part of the process - Standardization of the product design (common parts across product lines) NEW MARKET CREATION. Remanufactured products can’t be sold at the same price as a new one, therefore, lower prices are more attractive to consumer and a new problem arises. The cannibalization, which means that these lower priced products eat the new products market. Thus, there must be control to avoid it, although it is hard to measure it. Remanufactured stock’s unique problem: There is no control over the rate at which the returns flow because this rate depends on the rate of returned products. We cannot buy as much as we want of returned stock as we do with raw material. 9.2 Legislation 1. Take-Back Legislation Who should be responsible of the handling of waste? In Europe WEEE Directive of 2003 regulated it, but now it is 2019/12/EU WEEE Directive of July 4, 2012. In Japan it is the SHAR. And in the US, California is the only state where the consumers pay an ARF (Advanced Recycling Fee). Characteristics of the product: short life-cycle, expensive materials, expensive repairing. 2. Legislation of Greenhouse gas emissions CAP-AND-TRADE LEGISLATION (legislación sobre comercio de emisiones). For CO2 emissions in Europe and in California (2011) and SO2 emissions in the US (acid rain). Other legislation can be established besides cap-and-trade like emissions tax, command- and-control legislation, certificates (ISO, EMS)… In the EU there is the EU ETS (Régimen de comercio de derechos de emission de la UE). The system works on a cap-and-trade basis, which means that organisations are entitled to emit a certain amount of greenhouse gases, above which they need to buy allowances from other organisations that have not reached their limit. Each allowance, called an allowance (EUA), is equivalent to one tonne of carbon dioxide (CO2), the most common greenhouse gas, and companies can buy and sell them among themselves according to their needs. Example: If we assume that the limit is 10 tonnes of carbon (10tCO2), and that a company at the end of the year has emitted 6tCO2 in total, it can choose to keep the remaining 4tCO2 for its own future needs or sell it to another company that has emitted more than the allotted 10tCO2. At the end of each year, companies must have acquired enough allowances to cover all emissions, otherwise heavy penalties are imposed. COMMAND-AND-CONTROL VS. MARKET-BASED INSTRUMENTS SUSTAINABLE DEVELOPMENT GOALS (SDGS) GRI STANDARDS The GRI Standards enable any organization – large or small, private or public – to understand and report on their impacts on the economy, environment and people in a comparable and credible way, thereby increasing transparency on their contribution to sustainable development. In addition to companies, the Standards are highly relevant to many stakeholders - including investors, policymakers, capital markets, and civil society. CORPORATE SOCIAL RESPONSIBILITY (CSR) - How products and services affect people and the environment - Stakeholders have strong opinions about environmental, social, and ethical issues - Doing what’s right can be beneficial to all stakeholders SUSTAINABILITY - Meeting the needs of the present without compromising the ability of future generations to meet their needs - More than “going green” - Includes employees, customers, community, and company reputation TRIPLE BOTTOM LINE Consider the systems necessary to support the three Ps: People, Planet, and Profit - People: o Decisions affect people o Globalization and outsourcing complicate the task o Supplier selection and performance criteria are important o Materials must be safe for people - Planet: o Planet’s environment o Look for ways to reduce the environmental impact of operations o Overachieving objective is to conserve scarce resources o Carbon footprint (CO2) and Greenhouse gas emissions (GHG) o 3 Rs: Reduce, Re-use, Recycle (even Repair) - Profit: o Social and environmental sustainability do not exist without economic sustainability o Staying in business requires making a profit o Alternate measures of success include risk profile, intellectual property, employee morale, and company valuation o Social accounting can supplement financial accounting to support economic sustainability 9. Humanitarian Operations and Recap There are Development Operations (constantly helping, like fix setups of NGOs in Africa against hunger) and Emergency Operations. TYPES OF DISASTERS: Natural: o Major Natural Disasters: Flood, cyclone, drought, earthquake o Minor Natural Disasters: Cold wave, thunderstorms, heat waves, mud slides… Manmade: o Major Manmade Disasters: setting of fires, epidemics, deforestation, pollution, wars… o Minor Manmade Disasters: road/train accident, riots, food poisoning, industrial disasters/crisis… Geophysical disasters have remained fairly stable since 1970, while climate related disasters have greatly increased. DISASTER TIMELINE SUPPLY SOURCES PLAYERS / STAKEHOLDERS Donors & Recipient Delivering Agency People in need Suppliers Agency Global or local NGOs Government Local organizations NGOs Foundations or Governments Governments Companies Military Coordination and Collaboration is key. AidMatrix – portal to match donors and agencies. INTER-ORGANIZATIONAL COLLABORATION WHAT IS THE RESPONSIBILITY OF COMPANIES IN RESPONDING TO NATURAL DISASTERS? Companies can choose how much they want to engage, but in a disaster they are not viewed solely as providers of goods and services, but as members of the community Corporate Social Responsibility Responsibility towards employees Companies need to be involved in security and relief because “being dead is bad for business” REPUTATION RULES “Reputation Rules” are strategies for building your company’s most valuable asset (reputation) Empathy Commitment Expertise Transparency Example: WALMART’s response to hurricane Katrina Short-term Benefits of relief actions: - Direct benefits to the affected citizens - Media exposure as a promotional vehicle - Lessen the impact on their business and their facilities Long-term Benefits of relief actions: - Customers will remember and remain loyal to the brand - Workers loyalty to the firm - This relief activity triggered this new Walmart focus on sustainability CURRENT STATE OF ‘SCM‘ IN MOST NGOs SCM IN WALMART